Determination of Bearing Capacity of Open-Ended Piles in Sand

mentale loadlibration

It can alsoaftcapacity,e pile loading piletisfactory

Determination of Bearing Capacity of Open-Ended Pilesin Sand

Kyuho Paik1 and Rodrigo Salgado, M.ASCE2

Abstract: The bearing capacity of open-ended piles is affected by the degree of soil plugging, which is quantified by the increfilling ratio ~IFR!. There is not at present a design criterion for open-ended piles that explicitly considers the effect of IFR on pilcapacity. In order to investigate this effect, model pile load tests were conducted on instrumented open-ended piles using a cachamber. The results of these tests show that the IFR increases with increasing relative density and increasing horizontal stress.be seen that the IFR increases linearly with the plug length ratio~PLR! and can be estimated from the PLR. The unit base and shresistances increase with decreasing IFR. Based on the results of the model pile tests, new empirical relations for plug loadannulus load capacity, and shaft load capacity of open-ended piles are proposed. The proposed relations are applied to a full-scaltest performed by the authors. In this load test, the pile was fully instrumented, and the IFR was continuously measured durdriving. A comparison between predicted and measured load capacities shows that the recommended relations produce sapredictions.

DOI: 10.1061/~ASCE!1090-0241~2003!129:1~46!

CE Database keywords: Bearing capacity; Pile load tests; Sand.

l

n

s

r

g

oe

t

tanoadbutone.

totheted

e

ntaln-

nsts,en-atese

tionand

toaln ahe

ties

v

e

r

/

Introduction

When an open-ended pile is driven into the ground, a soil pmay develop within the pile during driving, which may prevent opartially restrict additional soil from entering the pile. It is knowthat the driving resistance and the bearing capacity of open-enpiles are governed to a large extent by this plugging effect.

Many design criteria for open-ended piles, based on field techamber tests or analytical methods, have been [email protected].,Klos and Tejchman 1977; Nishida et al. 1985; American Petleum Institute~API! 1991; Randolph et al. 1991; Jardine et a1998#. For example, in the case of API RP2A~1991!, which isgenerally used for offshore foundation design, the bearing capity of an open-ended pile can only be estimated for either the fucoring mode or the fully plugged mode of penetration. In practicmost open-ended piles are driven into sands in a partially plugmode. Stefanoff and Boshinov~1977! suggested the use of onedimensional plug analysis, in which the soil plug is treated asseries of horizontal thin discs and the force equilibrium conditiis applied to each disc, to calculate plug capacity of an opended pile.

There have been modifications of one-dimensional plug anasis to improve predictive accuracy, such as the introduction of

1Associate Professor, Dept. of Civil Engineering, Kwandong UniKangwon-do 215-800, South Korea~corresponding author!. E-mail:[email protected]

2Associate Professor, School of Civil Engineering, Purdue Univ., WLafayette, IN 47907-1284. E-mail: [email protected]

Note. Discussion open until June 1, 2003. Separate discussions mbe submitted for individual papers. To extend the closing date by omonth, a written request must be filed with the ASCE Managing EditThe manuscript for this paper was submitted for review and possipublication on July 23, 2001; approved on May 23, 2002. This papepart of theJournal of Geotechnical and Geoenvironmental Engineer-ing, Vol. 129, No. 1, January 1, 2003. ©ASCE, ISSN 1090-0241/200346–57/$18.00.

46 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGI

ugr

ded

ts,

o-l.

ac-llye,ed

-ann-

ly-he

concept of the wedged soil plug~Murff et al. 1990; O’Neill andRaines 1991; Randolph et al. 1991!. Many test results show thathe soil plug can be divided into a wedged plug zone andunwedged plug zone. While the wedged plug zone transfers lto the soil plug, the unwedged plug zone transfers no loadprovides a surcharge pressure on top of the wedged plug zHowever, it is not easy to apply the one-dimensional analysispractical cases, because of the sensitivity of the method tolateral earth pressure coefficient, which is not easily estima~Brucy et al. 1991; Leong and Randolph 1991!. De Nicola andRandolph ~1997! addressed this by proposing a profile of thlateral earth pressure coefficientK along the soil plug length.

An alternative design method can be based on the incremefilling ratio ~IFR!. The degree of soil plugging is adequately quatified using the IFR~Paikowsky et al. 1989; Paik and Lee 1993!defined as

IFR5DL

DD3100~%! (1)

whereDL5increment of soil plug length~L! corresponding to asmall incrementDD of pile penetration depthD ~see Fig. 1!. Thefully plugged and fully coring modes correspond to IFR50 and100%, respectively. A value of IFR between 0 and 100% meathat the pile is partially plugged. A series of model pile tesusing a calibration chamber, were conducted on model opended piles instrumented with strain gauges in order to investigthe effect of IFR on the two components of bearing capacity: baload capacity and shaft load capacity. Based on the calibrachamber test results, empirical relationships between the IFRthe components of pile load capacity are proposed. In orderverify the accuracy of predictions made using the two empiricrelationships, a full-scale static pile load test was conducted ofully instrumented open-ended pile driven into dense sand. Tpredicted pile load capacities are compared with the capacimeasured in the pile load test.

.,

st

ustne

or.bleis

1-

NEERING / JANUARY 2003

ns,tly,

thethe

lator

h aa

r todles

ately.t toesti-ear-and

Fig. 1. Definition of incremental filling ratio and plug length ratio

e

th

tTlte

.ln

ee

eh

ini-tra-

coneber

s ofmm,at aileec-

sizethisileses

Ac-uce

inary.

llyoadrdertru-aticders.

6.5

ger

of---esttly

e ofases

JOURNAL OF GEOTECHNI

l

ye

9

hehe

estsof

8,ed

s-ry

dt

of 1250 mm. In order to simulate various field stress conditiotwo rubber membranes, which can be controlled independenwere installed on the bottom and inside the lateral walls ofcalibration chamber. The consolidation pressure applied totwo rubber membranes was maintained constant by a regupanel throughout each pile test.

The soil samples were prepared by the raining method witconstant fall height. The falling soil particles passed throughsand diffuser composed of No. 8 and No. 10 sieves in ordecontrol flow uniformity and fall velocity. The soil samples haDR523, 56, and 90%. After sample preparation, the sampwere consolidated to the desired stress state during approxim30 h by compressed air transferred to the rubber membranes

Measurements made in calibration chambers are subjecchamber size effects. Many researchers have attempted tomate the chamber size needed for boundary effects on pile bing capacity or cone resistance to become negligible. ParkinLunne ~1982! suggested 50 times the cone diameter as the mmum chamber diameter for chamber size effect on cone penetion resistance to become acceptably small. Salgado et al.~1998!,based on cavity expansion analyses, found that 100 times thediameter was the minimum chamber diameter to reduce chamsize effects on cone resistance to negligible levels. Diameterthe chamber and test pile used in this study are 775 and 42.7respectively. The lateral and bottom boundaries are locateddistance equal to 18.2 pile radii from the pile axis and 23.0 pradii below the maximum depth reached by the pile base, resptively. Considering the results of the research on chambereffects mentioned above, the size of the chamber used instudy is not sufficiently large for chamber size effects on pbearing capacity to be neglected. The flexible boundary caulower radial stresses than those that would exist in the field.cordingly, the chamber tests done as part of this study prodlower pile load capacities than those that would be observedthe field. A correction for chamber size effects is then necessIt is discussed in a later section.

Model Piles and Test Program

Model Pile

An open-ended pile is generally driven into sands in a partiaplugged mode, and its bearing capacity is composed of plug lcapacity, annulus load capacity, and shaft load capacity. In oto separate pile load capacity into its components, an insmented double-walled pile was used in the testing. A schemdiagram of the pile is shown in Fig. 2. The model pile was maof two very smooth stainless steel pipes with different diameteIt had an outside diameter of 42.7 mm, inside diameter of 3mm, and length of 908 mm.

The wall thickness of the test piles used in this study is larthan those of piles typically used in practice. Szechy~1959!showed that the degree of soil plugging and bearing capacitytwo piles with different wall thicknesses do not differ in a significant way~with bearing capacity increasing only slightly with increasing wall thickness!; only driving resistance depends significantly upon the wall thickness. So the load capacity of the tpiles reported in this paper are probably larger, but only slighso, than what would be observed in the field.

Eighteen strain gauges were attached to the outside surfacthe inner pipe at nine different levels in order to measure the bload capacity~summation of plug and annulus load capacitie!

Soil Sample Preparation

Soil Properties

Han river sand, a subangular quartz sand, withD1050.17 mm andD5050.34 mm, was used for all the calibration chamber modpile tests. The test sand is classified as poorly graded~SP! in theUnified Soil Classification System, so the maximum dry densiof the sand is near the low end of the typical range for sands. Tmaximum and minimum dry unit weights of the sand were 15.8and 13.04 kN/m3, respectively.

A series of laboratory tests were conducted to characterizesand. The results from these tests are summarized in Table 1.internal friction angle of the sand and the interface friction angbetween the sand and steel were measured from direct shearunder normal stresses of 40–240 kPa. The peak friction anglesthe sand with relative densities of 23, 56, and 90% were 3438.2, and 43.4°, respectively, and the critical-state friction angwas 33.7°. The peak interface friction angles between the pile athe sand were 17.0, 17.5, and 18.4° forDR523, 56, and 90%,respectively, and the critical-state interface friction angle wa16.7°. This angle is lower than commonly reported values bcause the test pile was made of stainless steel pipe with a vsmooth surface.

Calibration Chamber and Sample Preparation

All model pile tests were conducted in soil samples preparwithin a calibration chamber with a diameter of 775 and a heig

Table 1. Soil Properties of Test Sand

Property Value

Coefficient of uniformityCu 2.21Coefficient of gradationCc 1.23

Maximum void ratioemax 0.986

Minimum void ratioemin 0.629

Minimum dry densitygd,min 13.04 kN/m3

Maximum dry densitygd,max 15.89 kN/m3

Specific gravityGs 2.64

Peak friction anglefpeak 34.8–43.4°

Critical-state friction anglefc 33.7°Peak interface friction angled 17.0–18.4°Critical-state interface friction angledc 16.7°

CAL AND GEOENVIRONMENTAL ENGINEERING / JANUARY 2003 / 47

y toca-

wasipe,

ithtes.

air peen

pip

ep

1itth

igesesnv

r-um-del

d ins ofere.ng toheingpilegime.wasoadt 19

mm,iedf thedeld by

untwtionn theith

48 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGIN

nipen

e,e.

a-

e. 3.tx-is-the

oid

this uncertainty, in this paper we use the base load capacitanalyze the test results instead of the plug and annulus loadpacities separately. The base load capacity of the test pileobtained from the upper strain gauges located on the inner pfor which the measured vertical loads reached a limit value~Fig.3!.

Test Program

Seven model pile tests were performed in dry soil samples wthree different relative densities and five different stress staEach test is identified by a symbol with three letters~H high, Mmedium,L low!, signifying the levels of the relative density, vetical and horizontal stresses of the sample, respectively. A smary of all model pile tests is presented in Table 2. Five mopile tests were conducted in dense samples withDR590% andfive different stress states. Two model pile tests were conducteloose and medium samples consolidated to a vertical stres98.1 kPa and horizontal stress of 39.2 kPa. The model piles wdriven by a 39.2 N hammer falling from a height of 500 mmDuring pile driving, the soil plug length and the pile penetratiodepth were measured at about 40 mm intervals, correspondin94% of the pile diameter, in order to calculate the IFR. Tchange in soil plug length during pile driving was measured usa ruler introduced through an opening at the top plate of the~see Fig. 2!. In order to measure the soil plug length, drivinoperations were suspended for no more than a minute each tStatic pile load tests were performed when the pile baselocated at depths of 250, 420, 590, and 760 mm. The pile ltests were continued until the pile settlement reached aboumm ~44% of the pile diameter!, at which point all the test pileshad reached a plunging limit state~Fig. 4!. The ultimate load ofeach test pile is defined as the load at a settlement of 4.27corresponding to 10% of the pile diameter. The total load applto the pile head was measured by a load cell, and settlement opile head was measured by two dial gauges. Details of the mopile, sample preparation, and test program have been describePaik and Lee~1993!.

Model Pile Test Results

Pile Drivability

Fig. 5~a! shows pile penetration depth versus hammer blow cofor all the test piles. As shown in the figure, the hammer blocount per unit length of penetration increases as pile penetradepth increases, since the penetration resistances acting obase and shaft of the piles during driving generally increase w

Table 2. Summary of Model Pile Test Program

Testindicator

Initialrelativedensity

~%!

Initialverticalstress~kPa!

Initialhorizontal

stress~kPa!

Initialearth

pressurecoefficient

HLL 90 39.2 39.2 1.0HML 90 68.6 39.2 0.6HHL 90 98.1 39.2 0.4HHM 90 98.1 68.6 0.7HHH 90 98.1 98.1 1.0MHL 56 98.1 39.2 0.4LHL 23 98.1 39.2 0.4

from the load transfer curve along the inner pipe. Two strgauges were also attached to the outside surface of the outein order to measure shaft load capacity. A gap of 4 mm betwthe outer pipe and the pile toe, which was sealed with silicoprevented the base load from being transferred to the outerThe outer pipe, therefore, experienced only the shaft load.

Many researchers have relied on linear extrapolation to srate the base load capacity into plug and annulus capacities~Paikand Lee 1993; Choi and O’Neill 1997; Lehane and Gavin 200!.Linear extrapolation would apply strictly only if the inside unfriction between the pile and soil plug were constant betweensecond lowest strain gauge and the pile base, as shown in FIn reality, the inside unit friction between the soil plug and the tpile increases dramatically near the pile base. Use of lineartrapolation, therefore, leads to an overestimation of annular retance. This overestimation increases as the distance betweelowest strain gauge and the pile base increases. In part to a

Fig. 3. Determination of plug and annulus loads

EERING / JANUARY 2003

Fig. 4. Load–settlement curves from model pile load tests

pl

eg.e

lee-he

gle

site

e,ilr

veige

ois

ana

JOURNAL OF GEOTECHNIC

e

d

tyst

t

Fig. 5. Driving test results:~a! hammer blow count, and~b! soil pluglength

ion

edpileng-se

LR.

penetration depth. The vertical stress applied to the soil samhad little effect on the cumulative blow count. However, the blowcount necessary to drive the pile to a certain depth decreasrapidly with decreasing horizontal stress. It is also seen in Fi5~a! that the blow count necessary for driving the pile to somrequired depth increases with increasing relative density.

Soil Plugging

The degree of soil plugging in an open-ended pile affects pibehavior significantly. The IFR is a good indicator of the degreof soil plugging. During the model pile tests, the IFR was measured at increments of 40 mm of penetration. The change of tsoil plug length with pile penetration depth is plotted in Fig. 5~b!.It is seen in the figure that the soil plug length developed durinpile driving increases as the horizontal stress of the soil sampincreases for the same relative density, and as the relative denincreases for the same stress. It can also be seen that everypile, during static load testing, advances in fully plugged modirrespective of the initial soil condition and the degree of soplugging during pile driving. The static load tests appear as shovertical lines in Fig. 5~b!, meaning that penetration depth in-creases while soil plug length remains unchanged.

Fig. 6 shows changes of IFR with soil state~relative density,vertical stress, and horizontal stress!. Fig. 6~a! shows IFR versusDR for tests withsv8598.1 kPa andK050.4. Fig. 6~b! shows IFRversussv8 for tests withDR590% andsh8539.2 kPa. Fig. 6~c!shows IFR versussh8 for DR590% andsv8598.1 kPa. It is ob-served that the IFR increases markedly with increasing relatidensity and with increasing horizontal stress. These changesIFR reflect the decreasing amount of compaction of the soil pluduring pile driving as the relative density and stress level in thsoil increase. However, the IFR is relatively insensitive tchanges in the vertical stress applied to the soil sample. Thmeans that the IFR of an open-ended pile would be higher foroverconsolidated sand than for a normally consolidated sandthe sameDR andsv8 .

Fig. 7 shows IFR versus plug length ratio~PLR! for the cham-ber test results and for the test results of Szechy~1959!; Klos and

n

t

Tejchman~1977!; Brucy et al.~1991!; and Paik et al.~2002!. ThePLR is defined as the ratio of soil plug length to pile penetratas ~see Fig. 1!

PLR5L

D(2)

In Fig. 7~b!, the data from Paik et al.~2002! were obtained froma full-scale pile with diameter of 356 mm driven into submergdense sands. The remaining data were obtained from modeltests using piles with various diameters driven into dry sand raing from loose to medium dense~the diameter of each test pile iindicated in the figure!. Fig. 7~a! shows that IFR, measured at thfinal penetration depth, increases linearly with increasing P

AL AND GEOENVIRONMENTAL ENGINEERING / JANUARY 2003 / 49

m-

eses

the

is

Fig. 6. Incremental filling ratio versus~a! relative density forsv8598.1 kPa andK050.4; ~b! vertical stress forDR590% andsh8539.2 kPa; and ~c! horizontal stress forDR590% and sv8598.1 kPa

50 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINE

The relationship between PLR and IFR for the calibration chaber tests can be expressed as follows:

IFR~%!5109•PLR222 (3)

This equation slightly underestimates the IFR for PLR valugreater than 0.8 and slightly overestimates it for PLR valulower than 0.7, as shown in Fig. 7~b!. In general, it is known thatthe IFR is a better indicator of the degree of soil plugging thanPLR ~Paikowsky et al. 1989; Paik and Lee 1993!. In the field,however, it is easier to measure the PLR than the IFR. Eq.~3! canbe used to estimate the IFR from the PLR, when only the PLRmeasured in the field.

Base and Shaft Load Capacities

The ultimate unit base resistanceqb,c measured in the calibrationchamber is plotted versus relative density~for sv8598.1 kPa andK050.4), versus vertical stress~for DR590% and sh8539.2 kPa) and versus horizontal stress~for DR590% and

Fig. 7. Plug length ratio versus incremental filling ratio~a! for cham-ber test results, and~b! for other test results

ERING / JANUARY 2003

asesitytoof

s a

withtive

for

g IFRwithse

ess,atepen-

ling

Fig. 8. Unit base resistance versus~a! relative density forsv8598.1 kPa andK050.4; ~b! vertical stress forDR590% andsh8539.2 kPa; and ~c! horizontal stress forDR590% and sv8598.1 kPa

JOURNAL OF GEOTECHNIC

sv8598.1 kPa) in Fig. 8. It is apparent that the ultimate unit bresistance increases significantly with increasing relative denand increasing horizontal stress, but is relatively insensitivevertical stress. This is consistent with experimental resultsBaldi et al. ~1981!; Houlsby and Hitchman~1988!; and Vipu-lanandan et al.~1989!, which showed that cone resistance wafunction of lateral effective stress.

Fig. 9 shows the ultimate unit base resistance, normalizedrespect to the horizontal stress, versus IFR for different reladensities, and the ultimate unit base resistance versus IFRdense sand. It can be seen in Figs. 9~a and b! that the ultimate unitbase resistance of open-ended piles increases with decreasinand that the rate of change of ultimate unit base resistanceIFR increases withDR . It is also seen that the ultimate unit baresistance increases with relative density at constant IFR.

Fig. 10 shows the ultimate unit shaft resistancef so,c measuredin the calibration chamber versus relative density, vertical strand horizontal stress. Similarly to what is observed for ultimunit base resistance, the ultimate unit shaft resistance of an o

Fig. 9. Normalized unit base resistance versus incremental filratio ~a! for sv8598.1 kPa andK050.4, and~b! for DR590%

AL AND GEOENVIRONMENTAL ENGINEERING / JANUARY 2003 / 51

talig.toities

ffer-cts.

uta-izetionlenthe

at ais-dedthethe

,reag,pilesterR.

ile,ed at

being,hisere

or-

erified

ly

Fig. 10. Unit shaft resistance versus~a! relative density forsv8598.1 kPa andK050.4; ~b! vertical stress forDR590% andsh8539.2 kPa; and ~c! horizontal stress forDR590% and sv8598.1 kPa

52 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINE

ended pile increases with both relative density and horizonstress, but is insensitive to the vertical stress. It is clear from F10~c! that the ultimate unit shaft resistance is linearly relatedthe horizontal stress. The ultimate base and shaft load capacof the test piles are listed in Table 3.

Correction of Chamber Test Results for ChamberSize Effects

Adjustment of Pile Diameter

Pile load capacities measured in a calibration chamber are dient from those measured in the field due to chamber size effeIn order to use the calibration chamber test results for comption of pile load capacity in the field, corrections for chamber seffects were performed for every chamber test. In the estimaof chamber size effects, the ratio of the chamber to the equivadiameter of the model pile used in the tests is required. Tequivalent diameter of an open-ended pile is the diameter thpile with solid cross-section would have to have in order to dplace the same soil volume during installation as the open-enpile. The equivalent diameter of open-ended piles varies withdegree of soil plugging, because the soil displacement aroundpile due to pile driving increases with decreasing IFR~Randolphet al. 1979!. For example, if a pile is driven in fully coring modethe equivalent pile diameter is calculated from an equivalent aequal to the annular area. If a pile is fully plugged during drivinthe gross cross-sectional area of the pile should be used. Fordriven in a partially plugged mode, the equivalent pile diamecan be determined through interpolation with respect to the IFThis is summarized, mathematically, as follows:

If IFR>100%, dp5A~d022di

2! (4a)

If IFR50%, dp5d0 (4b)

If 0%<IFR<100%,

dp5d02@d02A~d022di

2!#•IFR~%!

100(4c)

in which dp5equivalent pile diameter;d05outer pile diameter;anddi5inner pile diameter.

Considering the pile driving mechanism of an open-ended pthe base load capacity of the pile depends on the IFR measurthe final penetration depth. The shaft load capacity shouldrelated to the average value of the IFR measured during drivwhich is equal to the PLR at the pile penetration depth. In tstudy, therefore, the equivalent pile diameters for each test wcomputed for the base and shaft load capacities using Eqs.~4!.The IFR and PLR at the pile penetration depth are used for crection of the base and the shaft load capacity, respectively.

Field Pile Load Capacity

Salgado et al.~1998! conducted a theoretical analysis of chambsize effect for cone penetration resistance in sand and quantthe size effect as a function of soil state (DR andsh8) and chamberto pile diameter ratio. According to their results, which also appto displacement piles, the ratioqc,cc/qc,ff of chamber to field coneresistances for normally consolidated sands withDR523, 56,

ERING / JANUARY 2003

Table 3. Summary of Model Pile Test Results and Size Effect Factors

Testindicator

Testdepth~mm!

Soil pluglength~mm!

IFR~%! PLR

Base loadcapacity

~kN!

Shaft loadcapacity

~kN!

Size Effect Factor

Baseload

Shaftload

HLL 256 250 78.4 0.98 2.60 0.63 0.50 0.54420 366 71.4 0.87 2.91 0.90 0.49 0.51592 478 67.0 0.81 3.59 1.57 0.48 0.50760 571 54.4 0.75 3.91 2.13 0.46 0.49

HML 250 251 88.0 1.00 2.50 0.50 0.52 0.54420 373 76.3 0.89 2.85 0.81 0.50 0.52589 483 69.0 0.82 3.67 1.39 0.48 0.50760 583 57.4 0.77 4.30 2.23 0.47 0.49

HHL 250 251 84.2 1.00 2.42 0.53 0.51 0.54420 369 73.0 0.88 2.81 0.90 0.49 0.51590 477 69.5 0.81 3.54 1.65 0.48 0.50758 575 60.0 0.76 4.29 2.05 0.47 0.49

HHM 252 255 87.9 1.01 3.09 0.70 0.52 0.55420 381 78.6 0.90 3.57 1.45 0.50 0.52591 501 73.9 0.85 4.66 2.49 0.49 0.51761 614 72.1 0.81 4.91 3.60 0.49 0.50

HHH 251 266 92.6 1.06 4.53 1.36 0.53 0.56420 398 82.9 0.95 4.66 2.46 0.51 0.53590 521 79.8 0.88 5.40 3.93 0.50 0.52760 644 77.8 0.85 5.78 5.70 0.50 0.51

MHL 247 236 75.9 0.96 1.82 0.28 0.53 0.58419 347 67.4 0.83 2.17 0.49 0.51 0.55589 445 60.5 0.76 2.41 0.65 0.50 0.53757 532 53.9 0.70 2.82 1.00 0.49 0.52

LHL 247 224 71.1 0.91 1.01 0.18 0.61 0.66419 319 56.5 0.76 1.23 0.36 0.58 0.62581 401 52.4 0.69 1.46 0.59 0.57 0.60756 472 42.6 0.62 1.56 0.66 0.56 0.59

esehip

0.6llyit

and

esl

et

90%, and diameter ratio in the 10–45 range can be approximas

qc,cc

qc,ff5F1.08310223S Dc

dpD10.31G for DR523% (5a)

qc,cc

qc,ff5F1.02310223S Dc

dpD10.24G for DR556% (5b)

qc,cc

qc,ff5F7.79310233S Dc

dpD10.27G for DR590% (5c)

In these equations,qc,cc5cone resistance measured in a cabration chamber;qc,ff5field cone resistance; andDc /dp5ratio ofchamber to equivalent pile diameter. The chamber size effecttors for the base and shaft load capacities estimated by Eq.~5! arelisted in Table 3. The field pile load capacity can then be obtaiby dividing the chamber pile load capacity by the correspondsize effect factors.

New Design Equations for Load Capacity ofOpen-Ended Piles

Base Load Capacity

Fig. 11 shows the ultimate unit field base resistanceqb, f , normal-ized with respect to the horizontal effective stresssh8 at the pile

JOURNAL OF GEOTECH

ated

li-

fac-

neding

base, versus IFR for piles driven into sands with various relativdensities. The figure shows that the normalized unit field baresistance increases linearly with decreasing IFR. The relationsbetweenqb, f /sh8 and IFR can be expressed as

qb, f

ash85326– 295•

IFR~%!

100(6)

with a coefficient of determinationr 250.82. In this equation, thea values, a function of the relative density, were obtained fromthe calibration chamber tests as equal to 1.0 for dense sands,for medium sands, and 0.25 for loose sands. In the case of fuplugged piles~IFR50!, which behave as closed-ended piles, unfield base resistance is expressed asqb, f5326sh85130sv8 for nor-mally consolidated dense sands withK050.4. This is consistentwith the unit base resistance of a closed-ended pile in dense sproposed by the Canadian foundation engineering manual~CGS1992!. In order to predict base load capacity of open-ended pilusing Eq.~6!, it is necessary to know either the IFR or the soiplug length at the final penetration depth@from which the IFR canbe estimated through Eq.~3!#. A technique for measuring IFRduring pile installation will be described in a later section. Notthat Eq.~6! should be used only for piles driven into sands, nofor piles installed using vibratory hammers.

NICAL AND GEOENVIRONMENTAL ENGINEERING / JANUARY 2003 / 53

nta ntal

thein-

e o

ld;g;tionile

1.andensalseratimthenlnd.

tedsonit

e istest

t

iteFig

54 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINE

l Fig. 12. Normalized field unit shaft resistance versus incremefilling ratio

ing-

dedwn356asechede of

7.048.2lugtwoteelg,

d on. Atdi-

asleaftertestero-

bothoadereTheent

headloadtheike-

r andhead

shaftthethe

Shaft Load Capacity

The average ultimate field unit shaft resistancef so,f for the modelpiles, normalized with respect toK0sv8 tandc , is plotted versusPLR in Fig. 12 for various relative densities. It can be seen infigure that the normalized ultimate field unit shaft resistancecreases with decreasing PLR. The field unit shaft resistancpiles driven into dense sand can be expressed as follows:

f so,f

~K0sv8 tandc!b57.224.8•PLR (7)

in which f so,f5average ultimate unit shaft resistance in the fieK05lateral earth pressure coefficient before pile drivinsv85average vertical effective stress over the whole penetradepth;dc5critical-state interface friction angle between the pand the soil; andb5function of the relative density. Theb valueswere obtained from the calibration chamber tests as equal tofor dense sands, 0.4 for medium sands, and 0.22 for loose sIn the case of closed-ended piles in normally consolidated dsands withK050.4, the normalized unit shaft resistance equ7.2. This equation may be interpreted as implying that the latstress on the closed-ended pile driven in dense sands is 7.2higher than that before pile driving. This is consistent withlateral earth pressure coefficient ofK52 – 3, which the CanadiaFoundation Engineering Manual~CGS 1992! suggested for steepiles withd520° driven into a normally consolidated dense sa

Application of New Empirical Relations

Field Pile Load Test

A full-scale, field pile load test was performed on an instrumenopen-ended pile at Lagrange County in northern Indiana. Theat the site is gravelly sand with maximum and minimum dry uweights of 18.64 and 15.61 kN/m3, respectively. A 2.0 m thick filllayer was removed before pile driving. The groundwater tablat a depth of 3 m below the soil surface. Standard penetrationand cone penetration test results indicate that the first 3 m of thegravelly sand deposit are in a loose state (DR'30%), but the resof the deposit is in a dense to very dense state (DR'80%), asshown in Fig. 13. Note that the fill originally present at the swas removed before the piles were installed and tested, and

f

0s.e

les

il

.

13 accordingly does not include data for the fill. The resultoverconsolidation ratio~OCR! is also shown in Fig. 13 as a function of depth.

The test pile was an instrumented double-walled open-enpile, constituted of two pipes with different diameters, as shoin Fig. 14. The open-ended pile had an outside diameter ofmm and wall thickness of 32 mm. In order to measure the band shaft load capacities directly, 20 strain gauges were attato the outer surface of the inner pipe and 18 to the outer surfacthe outer pipe. The open-ended pile was driven to a depth ofm using a single acting diesel hammer with a ram weight of 1kN and a maximum hammer stroke of 3.12 m. The soil plength during pile driving was measured continuously usingdifferent weights, which were connected to each other by a swire ~Fig. 15!. The heavier weight rested on top of the soil pluand the lighter weight hanged outside the pile. A scale markethe outside of the pile allowed measurement of the plug lengththe final penetration depth, the IFR for the pile was 77.5%, incating a partially plugged condition, and the PLR was 0.82.

The load applied to the pile during the static load test wmeasured using a 2 MN load cell, and the settlement of the pihead was measured with two dial gauges. The residual loadspile driving and the loads induced at the base and shaft of thepile during the load test were independently measured by rezing the values of all strain gauges attached to the test pilebefore pile driving and at the start of the static load test. The lwas applied to the pile head in increments of 147 kN, which wdecreased to 49–98 kN as the pile approached the limit load.load after each increment was maintained until the pile settlemstabilized at less than 0.5 mm/h. The settlements at the pilewere measured at 5, 15, 35, 55, 75, 95, and 120 min for eachstep. When the settlement did not stabilize within 120 min,settlement was measured only after stabilization ensued. Lwise, strain values for the strain gauges attached to the inneouter pipes were measured after the settlement of the pilestabilized.

Static Load Test Results

Fig. 16 shows the load–settlement curves for the base andload capacities of the full-scale open-ended pile. As shown infigure, the shaft load capacity reached its limit value before

ERING / JANUARY 2003

Fig. 13. Cone penetration test and standard penetration test results and overconsolidation ratio profile at test site

errran

tiely

ale

sis-de-

ydea-

y.

final load step. The ultimate total and base load capacities walso determined as the loads at a settlement of 35.6 mm, cosponding to 10% of the pile diameter. The ultimate base and shload capacities not accounting for residual loads were 715 a310 kN, respectively. The ultimate base and shaft load capaciaccounting for residual loads were 886 and 139 kN, respectiveIn practice, it is difficult to account for residual loads. Residuloads are induced in every driven pile, but their magnitude d

m-sitth

Fig. 14. Schematic of full-scale test pile

JOURNAL OF GEOTECHNICA

Fig. 15. Measurement of soil plug length during pile driving

ee-ftds.

-

pends on several factors. The use of the unit base and shaft retance values that have been corrected for residual loads forsigning a different pile installed in a different sand site wouldrequire estimation of the residual loads for that pile. This is verdifficult to do in practice. Accordingly, we base our suggestedesign values of shaft and base resistances on the values msured without any correction for residual loads, as is customar

Comparison of Computed and Measured Capacities

The bearing capacity of the test pile was predicted using the epirical relationships suggested in this study. Since the soil depowas overconsolidated by removal of the fill layer, the lateral earpressure coefficientK0 was taken as~Mayne and Kulhawy 1982!

K05~12sinf!OCRsin f (8)

Saturated unit weights of the sand aregsat520.1 kN/m3 for theloose sand and 21.2 kN/m3 for the dense sand, respectively. The

L AND GEOENVIRONMENTAL ENGINEERING / JANUARY 2003 / 55

tal

the

e, as

gree

for

n aAt

es

osttionea-. Intioncity,pile

sesnde-asesoilfromreasewithpect

in-

ru-for

dersedasedsed

od ofduempile

DOTram.me

res

56 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINE

s

f so,f

~K0sv8 tandc!b57.224.8•PLR57.224.8~0.82!53.26

Qshaft5 f so,f•Aso53.26K0sv8 tandcb~pd0D !

53.26S~b iKoisv i8 Di !pd0 tandc

53.26~0.3363.411.03191.3!p~0.356!tan 22.2°

5312.9 kN

in which D5penetration depth of the pile. Thus, the ultimate toload capacity can be calculated as

Qtotal5Qbase1Qshaft5539.41312.95852.3 kN

The base and shaft load capacities predicted using Eqs.~6! and~7! were 75.4 and 100.9% of the ultimate values measured inpile load test, respectively. The predictedQtotal5852.3 kN is areasonably close, conservative estimate of the measured valushown in Fig. 17.

Summary and Conclusions

The bearing capacity of open-ended piles is affected by the deof soil plugging, which can be quantified through the IFR. Mdesign criteria for open-ended piles do not consider the variaof pile load capacity with IFR, and a standard technique for msuring IFR during pile installation has not yet been proposedthis study, model pile tests were conducted using a calibrachamber to investigate the effect of IFR on the pile load capaand new empirical relations between the two components ofload capacity~base and shaft load capacities! and IFR were pro-posed based on the results of model pile tests.

The results of model pile tests show that the IFR decreawith decreasing relative density and horizontal stress, but is ipendent of the vertical stress. It is also seen that the IFR increlinearly with the PLR, which is defined as the ratio of the splug length to pile penetration depth, and can be estimatedthe PLR. The unit base resistance shows a tendency to incwith decreasing IFR, and it does so at a rate that increasesrelative density. The unit shaft resistance, normalized with resto horizontal stress, increases with decreasing IFR and withcreasing relative density.

A full-scale pile load test was also conducted on a fully instmented open-ended pile driven into gravelly sand. The IFRthe pile was continuously measured during pile driving. In orto check the accuracy of predictions made with the propoequations, the equations were applied to the pile load test. Bon the comparisons with the pile load test results, the propoequations appear to produce satisfactory predictions.

Acknowledgments

The research presented in this paper was performed in a peri1 year spent by the first writer as a postdoctoral fellow at PurUniversity. The first writer is grateful for support received frothe Korea Science and Engineering Foundation. The fieldload test done as part of this research was supported by INand FHWA through the Joint Transportation Research ProgThe assistance of Dr. Junhwan Lee and Bumjoo Kim with soaspects of this research is appreciated.

References

American Petroleum Institute~API!. ~1991!. Recommended practice foplanning, designing and constructing fixed offshore platforms, 19th

mean particle size is 0.4 mm. The critical state friction anglethe sand obtained from triaxial compression tests isfc533.3°;the interface friction angle between the pile and sand is takedc52fc/3522.2°, which is adequate for typical pipe piles.the depth of the pile base, OCR51.41, andK0 results equal to0.55. Using Eq.~6!, the ultimate base load capacityQbasecan beobtained as

qb, f

ash85326– 295•

IFR~%!

1005326– 295•

77.5

100597.4

Qbase5qb, fAb597.4•ash8S p•d02

4 D597.4~1.0!~0.553101.2!~0.0995!5539.4 kN

The ultimate shaft load capacity can be computed using Eq.~7!.The b values used in the calculations are 0.3 for the first 3 m inloose sand and 1.0 for depth greater than 3 m indense sands. Thvariation ofK0 with OCR along the whole depth of the pile waconsidered in the calculations, which are summarized next

Fig. 17. Comparison of predicted with measured load capaciti

ERING / JANUARY 2003

E.

ivi-

n-

.,

.

n

e

.,

f.

y,

-

–

–

d

er.’’

Ed., America Petroleum Institute, Dallas, Dallas, API RP2A.Baldi, G., Bellotti, R., Ghionna, V., Jamiolkowski, M., and Pasqualini,

~1981!. ‘‘Cone resistance in dry N.C. and O.C. sands.’’Proc., ConePenetration Testing and Experience, Geotechnical Engineering Dsion, ASCE, NewYork, 145–177.

Brucy, F., Meunier, J., and Nauroy, J. F.~1991!. ‘‘Behavior of pile plug insandy soil during and after driving.’’Proc., 23rd Annual OffshoreTechnology Conf., Houston, 145–154.

Canadian Geotechnical Society~CGS!. ~1992!. Canadian foundation en-gineering manual, 3rd Ed., The Canadian Geotechnical Society, Vacouver, B.C., Canada.

Choi, Y., and O’Neill, M. W. ~1997!. ‘‘Soil plugging and relocation inpipe pile during earthquake motion.’’J. Geotech. Geoenviron. Eng123~10!, 975–982.

De Nicola, A., and Randolph, M. F.~1997!. ‘‘The plugging behavior ofdriven and jacked piles in sand.’’Geotechnique,47~4!, 841–856.

Houlsby, G. T., and Hitchman, R.~1988!. ‘‘Calibration chamber tests of acone penetration in sand.’’Geotechnique,38~1!, 39–44.

Jardine, R. J., Overy, R. F., and Chow, F. C.~1998!. ‘‘Axial capacity ofoffshore piles in dense north sea sands.’’J. Geotech. GeoenvironEng.,124~2!, 171–178.

Klos, J., and Tejchman, A.~1977!. ‘‘Analysis of behavior of tubular pilesin subsoil.’’ Proc., 9th Int. Conf. on Soil Mechanics and FoundatioEngineering, Tokyo, 605–608.

Lehane, B. M., and Gavin, G. K.~2001!. ‘‘Base resistance of jacked piptails in sand.’’J. Geotech. Geoenviron. Eng., ASCE,127~6!, 473–480.

Leong, E. C., and Randolph, M. F.~1991!. ‘‘Finite element analysis ofsoil plug response.’’Int. J. Numer. Analyt. Meth. Geomech.,15, 121–141.

Mayne, P. W., and Kulhawy, F. H. ~1982!. ‘‘ K0-OCRrelationships in soil.’’J. Geotech Eng. Div., Am. Soc. Civ. Eng108~6!, 851–872.

Murff, J. D., Raines, R. D., and Randolph, M. F.~1990!. ‘‘Soil plugbehavior of piles in sand.’’Proc., 22nd Offshore Technology Con,Houston, 25–32.

Nishida, H., Ohta, H., Matsumoto, T., and Kurihara, K.~1985!. ‘‘Bearing

JOURNAL OF GEOTECHNIC

capacity due to plugged soil on open-ended pipe pile.’’Proc., Jpn.Soc. Civ. Eng.,364, 219–227.

O’Neill, M. W., and Raines, R. D.~1991!. ‘‘Load transfer for pipe piles inhighly pressured dense sand.’’J. Geotech. Eng.,117~8!, 1208–1226.

Paik, K. H., and Lee, S. R.~1993!. ‘‘Behavior of soil plugs in open-endedmodel piles driven into sands.’’Marine Georesources Geotechnolog11, 353–373.

Paik, K. H., Salgado, R., Lee, J. H., and Kim, B. J.~2002!. ‘‘The behaviorof open- and closed-ended piles driven into sand.’’J. Geotech. Geoenviron Eng., in press.

Paikowsky, S. G., Whitman, R. V., and Baligh, M. M.~1989!. ‘‘A newlook at the phenomenon of offshore pile plugging.’’Mar. Geotech.,8,213–230.

Parkin, A. K., and Lunne, T.~1982!. ‘‘Boundary effect in the laboratorycalibration of a cone penetrometer for sand.’’Proc., 2nd EuropeanSymp. on Penetration Testing, Amsterdam, The Netherlands, 761768.

Randolph, M. F., Leong, E. C., and Houlsby, G. T.~1991!. ‘‘One-dimensional analysis of soil plugs in pipe piles.’’Geotechnique,41~4!,587–598.

Randolph, M. F., Steenfelt, J. S., and Wroth, C. P.~1979!. ‘‘The effect ofpile type on design parameters for driven piles.’’Proc., 7th EuropeanConf. on Soil Mechanics, British Geotechnical Society, London, 107114.

Salgado, R., Michell, J. K., and Jamiolkowski, M.~1998!. ‘‘Calibrationchamber size effects on penetration resistance in sand.’’J. Geotech.Geoenviron. Eng.,124~9!, 878–888.

Stefanoff, G., and Boshinov, B.~1977!. ‘‘Bearing capacity of hollow pilesdriven by vibration.’’ Proc., 9th Int. Conf. on Soil Mechanics anFoundation Engineering, Tokyo, 753–758.

Szechy, C. H.~1959!. ‘‘Tests with tubular piles.’’Acta Technica,24,181–219.

Vipulanandan, C., Wong, D., Ochoa, M., and O’Neill, M. W.~1989!.‘‘Modelling displacement piles in sand using a pressure chambFoundation engineering: Current principles and practices, Vol. 1,ASCE, New York, 526–541.

AL AND GEOENVIRONMENTAL ENGINEERING / JANUARY 2003 / 57